CN109142509B - Round steel magnetic powder flaw detection method and device - Google Patents
Round steel magnetic powder flaw detection method and device Download PDFInfo
- Publication number
- CN109142509B CN109142509B CN201710502185.XA CN201710502185A CN109142509B CN 109142509 B CN109142509 B CN 109142509B CN 201710502185 A CN201710502185 A CN 201710502185A CN 109142509 B CN109142509 B CN 109142509B
- Authority
- CN
- China
- Prior art keywords
- round steel
- identification
- image
- detected
- area
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N27/00—Investigating or analysing materials by the use of electric, electrochemical, or magnetic means
- G01N27/72—Investigating or analysing materials by the use of electric, electrochemical, or magnetic means by investigating magnetic variables
- G01N27/82—Investigating or analysing materials by the use of electric, electrochemical, or magnetic means by investigating magnetic variables for investigating the presence of flaws
- G01N27/83—Investigating or analysing materials by the use of electric, electrochemical, or magnetic means by investigating magnetic variables for investigating the presence of flaws by investigating stray magnetic fields
- G01N27/84—Investigating or analysing materials by the use of electric, electrochemical, or magnetic means by investigating magnetic variables for investigating the presence of flaws by investigating stray magnetic fields by applying magnetic powder or magnetic ink
Abstract
The invention discloses a round steel magnetic powder flaw detection method and a round steel magnetic powder flaw detection device. The round steel magnetic powder inspection method comprises the following steps: determining a detection area and dividing the detection area into a plurality of identification areas; coating magnetic suspension on the surface of the round steel to be detected, enabling the round steel to rotate and advance, irradiating the round steel to be detected by using ultraviolet rays, and continuously obtaining a surface image of the round steel to be detected in the detection area; preprocessing the surface image to obtain a defect image to be judged, which contains suspected surface defects, in each identification area; and determining whether the round steel to be detected has the surface defects or not according to the length-width ratio of the defect image to be judged and the frequency of the suspected surface defects appearing in each identification area. Round steel magnetic particle inspection device includes ultraviolet light source, image acquisition module and data processing module. The round steel magnetic powder flaw detection method and the round steel magnetic powder flaw detection device can reasonably divide a detection area by utilizing the characteristics of the defects of the round steel, accurately position the surface defects of the round steel and realize the automation of magnetic powder flaw detection.
Description
Technical Field
The invention relates to the technical field of flaw detection equipment, in particular to a round steel magnetic powder flaw detection method and device.
Background
With the increasingly fierce market competition and the increasing demands of users, the round steel blanks are usually subjected to magnetic powder inspection to ensure the surface quality of products.
At present, a common method for steel product enterprises to detect the magnetic powder of round steel is to irradiate a workpiece coated with magnetic suspension by using an ultraviolet lamp with certain brightness, then an operator judges and positions defects by observing fluorescent textures on the workpiece, and finally marks the defects for positioning and grinding in a subsequent grinding process.
In chinese patent CN201083697Y, a non-contact round steel online fluorescent magnetic powder flaw detector is disclosed. The flaw detector performs flaw detection by the method, and specifically, a round steel workpiece enters fluorescent magnetic powder flaw detection equipment in a spiral mode through a conveying roller way, is magnetized and sprayed with magnetic suspension at a magnetizing station, flaw detection personnel perform flaw detection observation through a fluorescent detection device, the defective workpiece is picked out and polished, and the defect-free workpiece is conveyed into a finished product warehouse. There are several problems with this approach: firstly, the operator needs to confirm the defect manually, the detection speed is slow, and the working efficiency is low; secondly, the labor intensity of operators is high, and the work content is monotonous and repeated, so that the missing rate is high; and thirdly, ultraviolet light in a working site causes physical damage to people working for a long time.
In recent years, machine vision technology has been rapidly developed, and manufacturers and scientific research institutions have also started to research the use of industrial cameras to assist operators or perform full-automatic defect recognition. Based on the robot vision technology, the Chinese patent CN103412042A discloses an image display axle magnetic particle flaw detector, and provides an axle magnetic particle flaw detector capable of using an image acquisition system, wherein the flaw detector can enable flaw detection processes and results to form images with uniform brightness and high resolution, can store, print, network and reproduce and playback, relieves flaw detection workers from a closed environment, and realizes standardization, digitization and networking of axle magnetic particle flaw detection operation processes and data management. However, the flaw detection equipment applying the automatic identification of the patent still needs operators to manually confirm the flaws and position the spray marks, and the comprehensive objective of automatic flaw detection is not completely realized.
Aiming at the problem that the flaw detection equipment and the flaw detection method in the prior art can not completely realize automatic flaw detection and identification, the invention provides a round steel magnetic powder flaw detection method and a round steel magnetic powder flaw detection device capable of automatically and accurately positioning flaws.
Disclosure of Invention
In order to solve the problems, the invention provides a round steel magnetic particle inspection method and a round steel magnetic particle inspection device, which can reasonably divide a detection area by using the characteristics of round steel defects, accurately position the surface defects of the round steel and realize the automation of magnetic particle inspection.
In order to achieve the purpose, the round steel magnetic powder inspection method comprises the following steps:
s1, determining a detection area for detecting the surface defects of the round steel to be detected, and dividing the detection area into a plurality of identification areas;
s2, coating the magnetic suspension on the surface of the round steel to be detected, enabling the round steel to rotate and advance, irradiating the round steel to be detected by ultraviolet rays, and continuously obtaining the surface image of the round steel to be detected in the detection area;
s3, preprocessing the surface image to obtain a to-be-determined defect image containing suspected surface defects in each identification area;
and S4, determining whether the round steel to be detected has surface defects or not according to the length-width ratio of the defect image to be determined and the occurrence frequency of the defect image to be determined in each identification area.
Further, the identification area is arranged in the detection area along the moving direction of the round steel to be detected.
And further, determining an identification area for identifying the surface defects, tracking the positions of the surface defects when the round steel to be detected is determined to have the surface defects, and identifying the surface defects in the identification area.
Further, the length of each recognition area is the same as the identification area.
Further, the width of the identification region located in the middle is the same as the width of the logo region, the widths of the other identification regions are gradually reduced from the identification region to both sides, and the minimum width of the identification region is equal to or greater than 1/5 of the width of the logo region.
Further, the preprocessing the surface image includes: firstly, filtering the surface image, and then determining a defect image to be judged by utilizing gray level binarization.
Further, when the defect image to be judged is determined by utilizing gray level binarization, weighting processing is carried out on a binarization threshold value.
Further, when the length-width ratio of the defect image to be judged is larger than 10 and the suspected surface defects continuously appear at least twice in each identification area, the round steel to be detected is determined to have the surface defects.
The round steel magnetic powder flaw detection device adopts the round steel magnetic powder flaw detection method, and comprises the following steps:
the ultraviolet light source is arranged opposite to the detection area and irradiates the surface of the round steel to be detected in the detection area;
the image acquisition module is arranged opposite to the detection area and used for acquiring a surface image of the round steel to be detected in the detection area;
and the data processing module is connected with the image acquisition module and determines whether the round steel to be detected has surface defects or not according to the surface image acquired by the image acquisition module.
The identification module is connected with the data processing module and is used for identifying the surface defects determined by the data processing module.
According to the round steel magnetic particle inspection method and device, the characteristics of the defects of the round steel are utilized, the detection area of the round steel to be detected is designed, the detection area is reasonably divided into a plurality of identification areas, the surface image of the round steel to be detected is acquired aiming at the detection area, whether the surface defects exist in the round steel to be detected in each identification area is determined according to the acquired surface image, the function of accurately positioning the defects can be realized without manual confirmation, the automation of the magnetic particle inspection of the round steel is realized, the production cost of the round steel is saved, the labor cost in production is reduced, the detection accuracy is high, and the time cost of production is saved.
Drawings
FIG. 1 is a flow chart of the round steel magnetic particle inspection method of the invention;
FIG. 2 is a schematic view of a detection region and an identification region in the present invention;
FIG. 3 is a schematic structural view of the round steel magnetic powder flaw detection apparatus of the present invention;
fig. 4 is a schematic diagram of a detection area and an identification area according to an embodiment of the present invention.
Detailed Description
The structure, operation, and the like of the present invention will be further described with reference to the accompanying drawings.
According to the round steel magnetic powder flaw detection method provided by the invention, an ultraviolet light source is used for irradiating the surface of the round steel to be detected, a shooting device is used for shooting a fluorescent surface image, the fluorescent surface image is transmitted to a data processing module, and the data processing module is used for processing the surface image, so that the defect condition of the surface of the round steel to be detected is determined and the identification is completed. The method mainly comprises three processes of identification area division, defect identification and defect identification of a detection area.
As shown in fig. 1, specifically, the round steel magnetic powder inspection method according to the embodiment of the present invention includes the following steps:
and S1, determining a detection area for detecting the surface defects of the round steel to be detected, and dividing the detection area into a plurality of identification areas. Because wait to detect round steel supplied materials, when magnetic particle inspection device, be in the state of rotation and forward movement all the time, consequently, for the processing degree of difficulty of the acquisition degree of difficulty of reduce cost, round steel surface image and the round steel surface image who acquires, need not to gather and wait to detect the holistic surface image of round steel, only need select a detection area can.
In addition, in the production process of round steel of iron and steel enterprises, most of surface defects are scratches along the axial direction, and defects in other directions can be ignored in mass production. After the defects are found, in view of facilitating the operation of the sharpening process, the flaw detection operator usually marks a long and narrow area containing the defects for the entire sharpening machine to pull out. Therefore, in the embodiment of the invention, the identification area where the surface defect of the round steel to be detected is located is only required to be determined, and the area is integrally marked, so that the requirement of a production field can be met. Therefore, before identification and marking, a plurality of virtual identification areas need to be marked in the detection area, then the marking area where the surface defect appears at last is determined, and the marking time and the marking are determined at last according to the position of the marking area and the moving mode and speed of the round steel to be detected.
Because it uses its axis to rotate on the limit to wait to detect the round steel, limit forward movement, in the embodiment of the invention, in order to discern the defect on waiting to detect the round steel surface more accurately, the regional arrangement of discernment needs to be synchronous with the removal mode of waiting to detect the round steel, consequently, the discernment region can be in the detection area along the moving direction setting of waiting to detect the round steel, the moving direction of waiting to detect the round steel is the direction that the arrow head is shown in figure 2. As can be seen from fig. 2, the identification regions may be respectively numbered a0(B0), a1, B1, … … An, Bn from the middle to both sides.
The length of the identification area is larger than the length of common surface defects and smaller than the typical grinding length of a grinding machine, and the length of the defect area marked by an operator in manual magnetic powder flaw detection can be generally referred to. Due to the influence of the curvature of the round steel, the area with the same width on the surface of the round steel to be detected is projected to the shooting device to deform, the area A0(B0) in the middle of the detection window is the widest, the typical width of the grinding machine can be taken, and the two sides of the grinding machine are narrowed in sequence.
Thus, the width of each recognition area can be calculated as follows:
W A0(B0) ＝D 1
wherein, W A0(B0) 、W A1 、W B1 、……W A(n-1) 、W B(n-1) 、W An 、W Bn The width of each identification region is A0(B0), A1, B1, … … A (n-1), B (n-1), An and Bn, D1 is the typical width of the grinding machine for one-time grinding, R is the radius of the round steel to be detected, and n is the number of the identification region (n is the number of the identification region)>0)。
S2, treat round steel surface coating magnetic suspension and make it rotate to advance, utilize ultraviolet irradiation to treat and treat the round steel, acquire the surface image of treating the round steel in the detection area in succession, owing to along with treating the removal of waiting to treat the round steel, acquire in succession and treat the surface image of treating the round steel through this detection area, just can not omit any department on the surface of treating the round steel.
S3, preprocessing the surface image to obtain the defect image to be judged containing the suspected surface defect in each identification area. When the surface images are preprocessed, each frame of surface image is preprocessed according to the time sequence of acquiring the surface images, so that whether the round steel to be detected has surface defects or not is judged in the following process.
In an embodiment of the present invention, the preprocessing the surface image includes: firstly, filtering the surface image, and then determining a defect image to be judged by utilizing gray level binarization.
The filtering process is performed on the surface image to remove noise in the image. In the embodiment of the present invention, a median filtering method may be adopted to perform filtering processing. Wherein the equation of the median filtering is:
g(x,y)＝med{f(x-i,y-j)}(i,j)∈P
in the formula, g (x, y) is the gray value of the pixel in the median filtering output image, f (x-i, y-j) is the gray value of the pixel in the median filtering input image, and P is the template window. The template window may be a window of various shapes and sizes, typically taking the shape of a 3 x 3 or 5 x 5 square window. Therefore, before performing the median filtering, if the size of the surface image does not match the template window, the surface image needs to be clipped first to obtain a size matching the template window, and then the filtering process is performed.
In the embodiment of the invention, the surface image gray scale can be subjected to binarization processing to remove noise, and a defect image to be determined is obtained. Because the ultraviolet illumination intensity of each identification area is different, the reflected fluorescence is also different, the middle area is usually bright, and the edge area is usually dark, therefore, when determining the defect image to be determined, the binary threshold value is weighted.
The processing formula for weighting the binarization threshold is as follows:
α A0(B0) ＝A
wherein alpha is A0(B0) 、α A1 、α B1 、……α An 、α Bn The binary threshold values of the recognition regions A0(B0), A1, B1, … … An and Bn; a is the binary threshold of the center recognition area A0(B0), and is preferably set to the middle of the camera gray scale range in consideration of making full use of the performance of the imaging device, adjusting the light intensity, the light source position, the camera light intensity, and the like. After the threshold of each recognition area is calculated, a point on each recognition area image with a brightness lower than the threshold may be set as a background, and a point higher than the threshold may be set as a to-be-determined trap image.
And S4, determining whether the round steel to be detected has surface defects or not according to the length-width ratio of the defect image to be determined and the occurrence frequency of the defect image to be determined in each identification area. And when the length-width ratio of the defect image to be judged is larger than 10 and the suspected surface defects continuously appear at least twice in each identification area, determining that the round steel to be detected has the surface defects.
In the actual production of the round steel, because the surface defects of the round steel are in a multi-bit long strip shape, the aspect ratio of the defect image to be judged is large, the ratio of the aspect ratio can be generally preset, and when the aspect ratio of the defect image to be judged is larger than the preset aspect ratio, the defect image to be judged can be determined as the surface defect. In the embodiment of the present invention, the ratio of the preset aspect ratio may be set to be greater than 10.
In the embodiment of the present invention, a formula of a determination condition for determining an aspect ratio of a defect image to be determined as a surface defect is:
wherein W is the length of the defect image to be judged, and H is the width of the defect image to be judged; beta is the ratio of the preset length-width ratio.
In order to enhance the anti-interference capability and reduce the false recognition, the surface defect can be tracked starting at the recognition area when the surface defect is recognized for the first time, and the defect can be confirmed again in the next recognition area. If the defect is confirmed in a plurality of identification areas, the defect can be finally determined. The number of the identification areas required for judging the surface defect is adjusted according to the requirement of sensitivity and the number of the partitions. Usually, the number of the identification areas needed for judging the surface defects is not suitable to be set to 1, so that more misjudgments exist, and therefore, at least 2 identification areas are needed; too many are not desirable, and in order to reduce the missing judgment, the maximum identification area required for judging the surface defects is not more than n + 1.
According to the moving mode and the moving speed of the round steel to be detected, the time when the surface defect appears in one recognition area and then appears in the next recognition area again can be calculated, namely the recognition period T1, and in order to improve the image processing efficiency and reduce the image processing cost, the T1 can also be used as the data acquisition period of the surface image. The identification period T1 is calculated as follows:
wherein: d1 is the typical width of sharpening machine one time, and R is the radius of waiting to detect the round steel, and V1 is round steel circumferential direction's linear velocity.
In general round steel production practice, a magnetic particle inspection operator usually marks an axial long-strip region containing defects for coping and positioning. Therefore, in the embodiment of the invention, an identification area for identifying the surface defect can be further determined, and when the round steel to be detected is determined to have the surface defect, the position of the surface defect is tracked, and the surface defect is identified in the identification area.
In summary, in the embodiment of the present invention, the length of each identification area may be set to be the same as the identification area. Because the width of each recognition area is reduced gradually to the recognition area of both sides by the recognition area that is located the axis of waiting to detect the round steel, consequently, the width that is located the recognition area on waiting to detect the round steel axis can set up to the same with the width of sign region. Because of the influence of the curvature of the round steel, the upper and lower edge identification areas are smaller in ultraviolet illumination intensity and projection area on the image acquisition unit, so that the identification areas do not need to be acquired too much, and the areas with stronger fluorescence reflection on the surfaces of the round steel to be detected can be covered, namely 1/5 which meets the requirement that the minimum width of the identification areas on the two sides of the identification areas is greater than or equal to the width of the identification areas. That is, the width of the An and Bn regions is not preferably smaller than 1/5, which is the width of the A0(B0) region.
In the embodiment of the invention, when the surface defect is judged to rotate to the identification area along with the round steel to be detected, the identification area can be subjected to overall marking. The marking area is located at the downstream of the identification area in the moving direction of the round steel to be detected, the specific position can be adjusted according to station limitation and the diameter of the round steel, and the marking area is generally located on the same central line with the identification area A0(B0) located in the middle and is at a distance D2 from the identification area A0 (B0). In view of the ease of tracking and marking surface defects, a surface defect that has been rotated several revolutions away from the identified region A0(B0) should fall well into the marking region. Then:
wherein m is a positive integer, represents the number of cycles that the round steel has rotated, can determine according to the space restriction of flaw detection operation station, and V2 is the speed that the round steel axially gos forward.
Then, when the surface defect is determined, the identification work may be performed after time T2 when the surface defect leaves the identification region Bn.
As shown in fig. 3, the round steel magnetic particle inspection device adopting the round steel magnetic particle inspection method of the invention comprises an ultraviolet light source 1, an image acquisition module 3 and a data processing module 4. Wherein, ultraviolet light source 1 sets up with the detection area is relative, and ultraviolet light source 1 can send the ultraviolet ray of certain intensity, shines the surface of waiting to detect round steel 2 in the detection area. Because the round steel 2 to be detected coated with the magnetic suspension is magnetized, the magnetic powder can be gathered to the defect, and after the ultraviolet light is irradiated, the fluorescence on the magnetic powder is excited, so that a fluorescence image can be formed on the surface of the round steel 2 to be detected. The image acquisition module 3 and the detection area are arranged relatively, specifically, can be arranged on the recognition station, captures the fluorescent image of the surface of the round steel 2 to be detected in the detection area, converts the fluorescent image into an array of images, and finally obtains the surface image. The data processing module 4 is connected with the image acquisition module 3, and after the surface image is transmitted to the data processing module 4, the data processing module 4 can process the surface image and identify defects according to the surface image, so that whether the surface defect exists in the round steel 2 to be detected is determined.
In the embodiment of the present invention, the surface defect detection device may further include an identification module 5, where the identification module 5 is connected to the data processing module 4, and identifies the surface defect determined by the data processing module 4. When the data processing module 4 is used, the surface defects can be tracked, and when the surface defects move to the indicating station, the marking module 5 is used for spraying marks.
In the embodiment of the present invention, the ultraviolet light source 1 may be an ultraviolet inspection lamp having a certain ultraviolet intensity. The image acquisition module 3 can select a CCD or CMOS industrial camera with stronger anti-interference capability. The data processing module 4 can be a processing unit such as an industrial personal computer, a PLC, a DSP and the like. The identification module 5 may be an industrial marking machine.
In one embodiment of the present invention, the ultraviolet light source 1 may use 4200 μ W/cm 2 The fixed ultraviolet flaw detection lamp. The image acquisition module 3 is an industrial CCD camera, and a fluorescence image excited after the ultraviolet irradiates the round steel 2 to be detected can be captured by a 1/1.8' CCD and a 25-frame/second monochromatic industrial camera. The data processing module 4 is an industrial control computer, and the surface image signal is transmitted to the industrial control computer. The industrial control computer performs data processing and defect recognition by the following method, and controls the marking module 5 to mark the surface defect at the marking station after the surface defect is determined. Wherein, the indication module is a label spraying machine.
First, the detection area is divided into 5 identification areas from top to bottom in the detection area as shown in fig. 4. The length of each identification area is taken to be 0.3 m. Since the typical width D1 of the sharpening machine used for one-time sharpening after the surface defect is determined in this embodiment is 0.05m, the radius R of the round steel 2 to be detected is 0.1m, and the width of each identification area can be calculated as follows:
W A0(B0) ＝D 1 ＝0.05m
from the above calculation, since the widths of the recognition areas a2, B2 are already close to 1/5 of the center recognition area a0(B0) due to the influence of the curvature of the round steel, it is not necessary to add more recognition areas.
Further, the step of identifying the defect in each identification area is as follows:
the surface image is firstly subjected to median filtering to remove noise, and the formula is as follows:
g(x,y)＝med{f(x-i,y-j)}(i,j)∈P
where P here takes the form of a 3 x 3 square window.
And carrying out binarization processing on the surface image gray level to obtain a defect image to be judged. Because the ultraviolet illumination intensity of each identification area is different, the reflected fluorescence is also different, the middle area is usually bright, the edge area is usually dark, and the binary threshold value can be weighted. The gray scale of the industrial camera used in the embodiment of the invention is 0-127, and the threshold of the identification area A0(B0) needs to be set to be about 63 in consideration of fully utilizing the camera performance. The method comprises the steps of adjusting the illumination intensity, the light source position, the camera light flux amount and the like before flaw detection is started until the acquisition value of a fluorescence brightness camera of typical surface defects in a central recognition area is obviously higher than a threshold value 63; meanwhile, the acquisition value of the background fluorescence brightness camera on the surface of the round steel 2 to be detected is obviously lower than 63. The threshold for each identified region is as follows:
α A0(B0) ＝A＝63
after the threshold value of each identification area is calculated, the point with the brightness lower than the threshold value on the image of the identification area is assigned with 0, namely, the point is identified as the background, the part higher than the threshold value is assigned with 127, namely, the point is identified as the to-be-determined trap image, namely, the image is identified as the to-be-determined trap image
In the formula, h (x, y) is the assignment of the pixel in the binary output image, k (x, y) is the gray value of the pixel in the binary input image, and alpha i For each region threshold, i ∈ [ A ] 2 ,A 1 ,A 0 (B 0 ),B 1 ,B 2 ]。
Since most of the surface defects of the round steel 2 to be detected are long linear, the aspect ratio of the defect image to be determined should be relatively large, and in the embodiment of the present invention, the aspect ratio may be set to be 10, that is, the aspect ratio is set to be 10
When the aspect ratio of the defect image to be determined is greater than 10, the defect image can be identified as a surface defect.
The tracking of the surface defect is started when the surface defect is first identified in the a 0-a 2 regions, and after one identified region period, the surface defect is confirmed again in the next identified region. When the linear speed of circumferential rotation of the round steel 2 to be detected is V1 and is 0.2m/s, the identification period T1 is calculated as follows:
since the total number of the recognition areas is 5 in the present embodiment, considering that the surface defect recognized at the latest in the a0 area passes through 3 recognition areas at most, the number of the recognition areas required for determining the surface defect should be 3 or less; in addition, for less erroneous determination, the number of identification regions required for determining the surface defect should preferably be larger than 1, and is therefore taken as 2 here.
After the surface defects are recognized, the size of the marking region may be set to coincide with the central recognition region a0 (B0). The identification region is located downstream of the recognition region, specifically downstream of the central recognition region a0(B0) on the same centerline, at a distance from the recognition region a0(B0) calculated as follows:
in view of the space limitations of the stations, m in the formula is 4, and V2 is 0.2 m/s.
When a surface defect is identified in at least two of the identified regions, the surface defect can be finalized, and when the surface defect leaves the identified region Bn, the marking operation can be performed after time T2.
Wherein n is 2.
In conclusion, the round steel magnetic powder inspection device and the round steel magnetic powder inspection method provided by the embodiment of the invention can reasonably divide the inspection area, realize automatic identification and defect identification of online magnetic powder inspection, improve inspection capability and effect, and reduce labor intensity of manual inspection.
The foregoing is merely illustrative of the present invention, and it will be appreciated by those skilled in the art that various modifications may be made without departing from the principles of the invention, and the scope of the invention is to be determined accordingly.
Claims (7)
1. A round steel magnetic powder flaw detection method is characterized by comprising the following steps:
s1, determining a detection area for detecting the surface defects of the round steel to be detected, and dividing the detection area into a plurality of identification areas;
s2, coating a magnetic suspension on the surface of the to-be-detected round steel, enabling the to-be-detected round steel to rotate and advance, irradiating the to-be-detected round steel with ultraviolet rays, and continuously obtaining a surface image of the to-be-detected round steel in the detection area;
s3, preprocessing the surface image to obtain a to-be-determined defect image containing suspected surface defects in each identification area;
s4, determining whether the round steel to be detected has surface defects or not according to the length-width ratio of the defect image to be determined and the occurrence frequency of the defect image to be determined in each identification area;
the method further comprises the steps of determining an identification area for identifying the surface defects, tracking the positions of the surface defects when the round steel to be detected is determined to have the surface defects, and identifying the surface defects in the identification area; the width of the identification region in the middle is the same as that of the identification region, the widths of the other identification regions are gradually reduced from the identification region to both sides, and the minimum width of the identification region is greater than or equal to 1/5 of the width of the identification region.
2. The round steel magnetic particle inspection method according to claim 1, wherein the identification area is provided in the detection area along a moving direction of the round steel to be inspected.
3. The round steel magnetic particle inspection method according to claim 1, wherein the length of each of the identification regions is the same as the length of the identification region.
4. The round steel magnetic particle inspection method of claim 1, wherein the preprocessing the surface image comprises: and filtering the surface image, and determining the defect image to be judged by utilizing gray level binarization.
5. The round steel magnetic particle inspection method according to claim 4, wherein when the defect image to be determined is determined by means of gray level binarization, a binarization threshold value is subjected to weighting processing.
6. The round steel magnetic particle inspection method according to claim 1, wherein the presence of the surface defect in the round steel to be inspected is determined when the aspect ratio of the defect image to be inspected is larger than 10 and the suspected surface defect appears at least twice in each of the identification regions.
7. A round steel magnetic particle testing apparatus employing the round steel magnetic particle testing method according to any one of claims 1 to 6, comprising:
the ultraviolet light source is arranged opposite to the detection area and irradiates the surface of the round steel to be detected in the detection area;
the image acquisition module is arranged opposite to the detection area and used for acquiring a surface image of the round steel to be detected in the detection area;
the data processing module is connected with the image acquisition module and is used for determining whether the round steel to be detected has surface defects or not according to the surface image acquired by the image acquisition module;
the device also comprises an identification module, wherein the identification module is connected with the data processing module and is used for identifying the surface defects determined by the data processing module.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710502185.XA CN109142509B (en) | 2017-06-27 | 2017-06-27 | Round steel magnetic powder flaw detection method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710502185.XA CN109142509B (en) | 2017-06-27 | 2017-06-27 | Round steel magnetic powder flaw detection method and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109142509A CN109142509A (en) | 2019-01-04 |
| CN109142509B true CN109142509B (en) | 2022-08-16 |
Family
ID=64805165
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710502185.XA Active CN109142509B (en) | 2017-06-27 | 2017-06-27 | Round steel magnetic powder flaw detection method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109142509B (en) |
Families Citing this family (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110095527B (en) * | 2019-04-30 | 2023-12-15 | 冶金自动化研究设计院 | Automatic fluorescent magnetic powder flaw detection device and method on assembly line |
| CN113506223A (en) * | 2020-03-23 | 2021-10-15 | 上海西门子医疗器械有限公司 | Image defect detection method and device, storage medium and medical imaging system |
| CN114076794A (en) * | 2020-08-19 | 2022-02-22 | 宝山钢铁股份有限公司 | Automatic detection device and detection method for near-surface defects of small strip steel square billet |
| CN113358739A (en) * | 2021-06-30 | 2021-09-07 | 常州工学院 | Magnetic particle inspection detection device based on binocular image fusion technology |
| CN113702391B (en) * | 2021-09-08 | 2022-08-23 | 东力智能科技有限公司 | Method and device for compositely detecting defects on surface and near surface of steel billet |
| CN115980176B (en) * | 2023-03-20 | 2023-06-23 | 天津市特种设备监督检验技术研究院(天津市特种设备事故应急调查处理中心) | Spherical tank quality data analysis processing method and system based on magnetic powder detection |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP3394653B2 (en) * | 1996-07-15 | 2003-04-07 | 新日本製鐵株式会社 | Magnetization inspection method and apparatus |
| CN201527410U (en) * | 2009-11-05 | 2010-07-14 | 江苏融泰石油科技股份有限公司 | Continuous magnetic powder flaw detection device for drill collar |
| CN105092593A (en) * | 2015-08-27 | 2015-11-25 | 张小磊 | Steel plate defect detection method based on high-intensity illumination |
| CN105352977B (en) * | 2015-11-06 | 2018-09-04 | 苏州磁星检测设备有限公司 | Automatic flaw detection detection line and detection method |
-
2017
- 2017-06-27 CN CN201710502185.XA patent/CN109142509B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN109142509A (en) | 2019-01-04 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109142509B (en) | Round steel magnetic powder flaw detection method and device | |
| CN109141232B (en) | Online detection method for disc castings based on machine vision | |
| CN109115785B (en) | Casting polishing quality detection method and device and use method thereof | |
| CN102621156B (en) | Image-processing-based automatic micro part sorting system | |
| CN107290347B (en) | Automatic honeycomb carrier defect detection method | |
| CN106802215A (en) | A kind of device for detecting water leakage of water pipe and detection method | |
| Liu et al. | Automatic detection technology of surface defects on plastic products based on machine vision | |
| CN105973912A (en) | Leather surface defect detecting system and method based on machine vision | |
| CN112881427B (en) | Electronic component defect detection device and method based on visible light and infrared thermal imaging | |
| CN110728657A (en) | Annular bearing outer surface defect detection method based on deep learning | |
| CN106353336A (en) | Lens coating automatic detection system | |
| CN106529551B (en) | Intelligent recognition counting detection method for round-like objects in packaging industry | |
| Liu et al. | An automatic system for bearing surface tiny defect detection based on multi-angle illuminations | |
| CN112893172A (en) | Gasket size detection system and method based on machine vision, processing terminal and medium | |
| CN107891012B (en) | Pearl size and circularity sorting device based on equivalent algorithm | |
| Fu et al. | Medicine glass bottle defect detection based on machine vision | |
| CN116309277A (en) | Steel detection method and system based on deep learning | |
| CN111833350B (en) | Machine vision detection method and system | |
| CN113177925A (en) | Method for nondestructive detection of fruit surface defects | |
| CN111598831A (en) | Detection method and system for cylindrical roller bearing | |
| CN109622404B (en) | Automatic sorting system and method for micro-workpieces based on machine vision | |
| Giesko et al. | Advanced mechatronic system for in-line automated optical inspection of metal parts | |
| CN115861225A (en) | Bearing roller surface damage detection method based on damaged area fusion transformation | |
| CN109374628A (en) | The defect confirmation method of gravure printing roller surface defect intelligent measurement | |
| CN109632815A (en) | Surface defect on-line detecting system and method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |